Electrostatic phenomena are ubiquitous in biological processes such as protein folding, binding, and catalysis. Our current knowledge of electrostatic effects on protein stability is mainly derived from protein engineering experiments and theoretical studies using static-structure based Poisson-Boltzmann calculations. However, while macroscopic measurements often can not isolate electrostatic effects from others, the accuracy of theoretical predictions is limited by the lack of explicit treatment of protein dielectric response, conformational dynamics and effects due to residual structures in the unfolded state. As a result, despite two decades of research, important questions such as how and to what extent electrostatic interactions modulate protein stability have not been adequately answered. The lack of accurate means to predict electrostatic contributions not only hampers fundamental understanding of protein stability but also poses a roadblock for advancing computational protein design. The specific aims of this application are 1) to advance atomic-level studies of pH-dependent phenomena by further developing continuous constant pH molecular dynamics and related methodologies, and 2) to improve quantitative prediction and detailed understanding of electrostatic modulation of protein stability by studying several model systems including the N-terminal domain of ribosomal L9 protein, villin headpiece subdomain, leucine zipper, and meso-, thermo- and hyperthermophilic variants of peripheral subunit binding domain. The proposed method development will provide the community with powerful tools for studying a wide range of electrostatic phenomena in biology. The insights gained in the application studies are expected to shift the native-centric paradigm of protein stability and function and transform the static-structure based view of protein electrostatics. They will also help establish general principles for computational protein design.